Over the past few decades, there have been significant advancements in the field of wireless communication. Wireless technology has found applications in a variety of areas such as telephony, control of industrial devices, entertainment and many more. Some common examples of wireless communication systems include mobile phones, cordless phones, pagers, and wireless LANs.
Wireless communication systems typically involve the use of transmitters and receivers for the transmission and the reception of data signals respectively. The data signals are embedded in a carrier wave. The carrier wave is typically a sinusoid whose oscillation frequency is referred to as the carrier frequency. The carrier wave is modulated at the transmitter end according to certain characteristics of the data signal. During modulation, a particular characteristic, such as amplitude or frequency, of the carrier wave is varied according to the data signal. The carrier wave, which is modulated using the data signals, is termed as a modulated carrier wave. The modulated carrier wave is demodulated at the receiver end to recover the original data signals. Thus, the data signals are exchanged between the transmitter and the receiver.
For effective exchange of data signals in a wireless communications system, it is imperative that both the transmitter and the receiver operate at the same carrier frequency. Consistency of frequency is ensured by using frequency references in devices such as transmitters and receivers. A frequency reference is an oscillator that produces a standard frequency, from which the one rating frequencies of the receiver and the transmitter are derived. Typically, a frequency reference is implemented using a piezo-electric crystal. Other types of frequency references, including those constructed of integrated circuit elements such as resistors, inductors and capacitors, tend to be less accurate and less expensive than the crystal-based references. In general, the reference elements in the transmitter and the receiver are different and produce slightly different frequencies. Even if the transmitter and the receiver reference elements are of similar design, they may produce different frequencies or vary over time due to manufacturing variation and environmental factors such as temperature, vibration, and aging. This leads to a mismatch between the carrier frequencies at which the transmitter and the receiver operate. This mismatch is termed as frequency offset. The frequency offset between the transmitter and the receiver is a major hindrance in achieving efficient exchange of data signals between the two devices. The frequency offset needs to be compensated in order to improve the quality of wireless communication.
The usage of crystal frequency references at both the transmitter as well as the receiver ends presents an adequate technical solution to the problem of controlling the frequency offset. However, the large size and high cost of piezo-electric crystals are undesirable. The high cost of frequency reference is a limitation especially in low-cost, battery-powered communications equipment designed for sensing and logistics applications.
There exist a number of techniques dealing with frequency offset in wireless communication. In one known technique, an initial estimate of the frequency is obtained using a time average of either phase or correlation samples of a data packet preamble, the data packet being a collection of a fixed number of data bits. This method also estimates the frequency during the transmission of the actual data packet. Hence, an estimate of the frequency offset is obtained.
Another known technique teaches a method for automatically correcting the frequency at the receiver in order to nullify the frequency offset during the exchange of data bits. The frequency offset is estimated using a training sequence. The receiver has the knowledge of the training sequence but not that of the data bits. The frequency correction is achieved by back-rotating the received data bits by the frequency offset estimate. The training sequence is also back rotated by the frequency offset estimate. A second estimate of the frequency offset is obtained using the corrections in the frequency of the data bits as well as the training sequence. The second frequency offset estimate is used to further correct the frequencies of the data symbols.
An automatic frequency compensation apparatus is also known. This apparatus determines a correlation between a baseband signal and a predetermined signal. The baseband signal is the actual data signal that is transmitted. When the correlation exceeds a reference value, the apparatus effects the frequency offset compensation to the baseband. An equalizer is used after the frequency offset compensation. The frequency offset can further be predicted using the output of the equalizer. The prediction is also effected during the training period when the correlation exceeds the reference value.
While the above cited techniques and apparatus deal with the frequency offset problem, the receivers described in these publications require a large bandwidth. The requirement for a large bandwidth exists because the receivers have to process the received signal and also a range of offset frequencies. This results in an undesirable degradation in the receiver sensitivity.
There also exist other methods for achieving frequency offset compensation. For example, in Direct Sequence Spread Spectrum (DSSS) type of modulation, Differential Chip Detection (DCD) can mitigate frequency offset of the order of 10 to 30 percent of chip rate. DSSS is a modulation technique utilizing a digital code sequence having a coded bit rate, or chip rate, much higher than the information signal bit rate. Data signals are transmitted as a collection of a number of data bits. Each data bit consists of a number of code bits, or chips. In DCD, each chip is processed differentially with respect to previous chips. The frequency offset can be mitigated to a certain extent because the time difference between the chips is short. While this technique is more efficient in trading the frequency offset tolerance for receiver sensitivity than other techniques, there is a trade-off in the bandwidth of the spread signal that must be applied to mitigate a given amount of frequency offset. Further, the signal bandwidth is often constrained because of regulatory considerations. Thus, this technique is limited in its effectiveness in mitigating the frequency offset by the available bandwidth of the signal. Also, the higher chip rates imply increased power and complexity, which is undesirable in low-cost, battery powered communications hardware.
From the above discussion, it is evident that there exists a need for a technique that compensates large frequency offsets while minimizing degradation in the receiver sensitivity. The technique should minimize the power dissipation and the complexity in the receiver and the transmitter. The technique should mitigate the frequency offset but should not increase the size and cost of the receiver and the transmitter.